*Rakesh Gupta, Hemant Kumar Sharma, Manvendra Jaiswal, Rajeev Sharma, Narendra Nyola, Dr. Rajesh Yadav
Alwar Pharmacy College, M.I.A. Alwar,
Rajasthan, India 301001
Experimental design is a planned interference in the natural order of events by the researcher. A selected condition or a change (treatment) is introduced. Observations or measurements are planned to illuminate the effect of any change in conditions. Complex designs, usually involving a number of "control groups," offer more information than a simple group design. It involves the Experimental Design and Data Analysis. The various type of experimental design, e.g. Statistical(Randomized Blocks, Latin Square, Factorial Design), Quasi Experimental (Time Series, Multiple Time Series), True Experimental(Pretest-Posttest Control Group, Post-test: Only Control Group, Solomon Four-Group), Pre-experimental (Static Group, One Group Pretest-Posttest, Experimental One-Shot Case Study).Process Models for DOE is common to begin with a process model of the `black box' type (Quadratic model & Linear model).Full factorial designs in two levels, Full factorial designs not recommended for 5 or more factors. Replication provides information on variability, Factor settings in standard order with replication, No randomization and no center points; Randomization provides protection against extraneous factors affecting the results. Contour plot Display 3-d surface on 2-d plot Vertical axis, Horizontal axis, Lines CCD designs start with a factorial or fractional factorial design (with centre points) and add "star" points to estimate curvature; A CCD design with k factors has 2k star points, 3 types of CCD designs, which depend on where the star points are placed Circumscribed (CCC), Inscribed (CCI), Face Cantered (CCF); the value of α is chosen to maintain rotatability.